Automatic Phase-Picking Method for Detecting Earthquakes Based on the Signal-to-Noise-Ratio Concept

摘要

The automatic phase-picking detection of earthquakes is a challenge under the background of big data and strong noise circumstances. The short-term average/long-term average (STA/LTA) ratio is widely used to detect earthquake due to its simplicity and robustness. However, STA/LTA-based methods may not perform well with noisy data. Based on the signal-to-noise-ratio (SNR) concept, a short-term power/long-term power (STP/LTP) ratio method is proposed. The characteristic function and the detection thresholds of the STP/LTP method are given physical meanings. Through a sample analysis, the STP/LTP detection results of both the P and S phases are better than the results of the STA/LTA by means of mean deviation, standard deviations, distributions of detection results, error rate, and missed rate on different SNR levels. In general, the STP/LTP method inherits the simple characteristics of the STA/LTA method, and it is suitable for phase picking of low-SNR seismic data.